Evaluation of a community health worker intervention and theWorld Health Organization’s Option B versus Option A to improve

antenatal care and PMTCT outcomes in Dar es Salaam,Tanzania: study protocol for a cluster-randomized controlled

health systems implementation trial

(Article begins on next page)

The Harvard community has made this article openly available.Please share how this access benefits you. Your story matters.

Citation Sando, D., P. Geldsetzer, L. Magesa, I. A. Lema, L. Machumi,M. Mwanyika-Sando, N. Li, et al. 2014. “Evaluation of acommunity health worker intervention and the World HealthOrganization’s Option B versus Option A to improve antenatalcare and PMTCT outcomes in Dar es Salaam, Tanzania: studyprotocol for a cluster-randomized controlled health systemsimplementation trial.” Trials 15 (1): 359. doi:10.1186/1745-6215-15-359. http://dx.doi.org/10.1186/1745-6215-15-359.

Published Version doi:10.1186/1745-6215-15-359

Accessed February 17, 2015 5:47:18 AM EST

Citable Link http://nrs.harvard.edu/urn-3:HUL.InstRepos:13454680

Terms of Use This article was downloaded from Harvard University's DASHrepository, and is made available under the terms and conditionsapplicable to Other Posted Material, as set forth athttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA

TRIALSSando et al. Trials 2014, 15:359http://www.trialsjournal.com/content/15/1/359

STUDY PROTOCOL Open Access

Evaluation of a community health workerintervention and the World Health Organization’sOption B versus Option A to improve antenatalcare and PMTCT outcomes in Dar es Salaam,Tanzania: study protocol for a cluster-randomizedcontrolled health systems implementation trialDavid Sando1,2†, Pascal Geldsetzer1†, Lucy Magesa2, Irene Andrew Lema2, Lameck Machumi2,Mary Mwanyika-Sando2, Nan Li1, Donna Spiegelman3, Ester Mungure2, Hellen Siril2,4, Phares Mujinja4,Helga Naburi4,5, Guerino Chalamilla2, Charles Kilewo5, Anna Mia Ekström6,7, Wafaie W Fawzi1 andTill W Bärnighausen1,8*

Abstract

Background: Mother-to-child transmission of HIV remains an important public health problem in sub-Saharan Africa. AsHIV testing and linkage to PMTCT occurs in antenatal care (ANC), major challenges for any PMTCT option in developingcountries, including Tanzania, are delays in the first ANC visit and a low overall number of visits. Community healthworkers (CHWs) have been effective in various settings in increasing the uptake of clinical services and improvingtreatment retention and adherence. At the beginning of this trial in January 2013, the World Health Organizationrecommended either of two medication regimens, Option A or B, for prevention of mother-to-child transmission of HIV(PMTCT). It is still largely unclear which option is more effective when implemented in a public healthcare system. Thisstudy aims to determine the effectiveness, cost-effectiveness, acceptability, and feasibility of: (1) a community healthworker (CWH) intervention and (2) PMTCT Option B in improving ANC and PMTCT outcomes.

Methods/Design: This study is a cluster-randomized controlled health systems implementation trial with a two-by-twofactorial design. All 60 administrative wards in the Kinondoni and Ilala districts in Dar es Salaam were first randomlyallocated to either receiving the CHW intervention or not, and then to receiving either Option B or A. Under the standardof care, facility-based health workers follow up on patients who have missed scheduled appointments for PMTCT, firstthrough a telephone call and then with a home visit. In the wards receiving the CHW intervention, the CHWs: (1) identifypregnant women through home visits and refer them to antenatal care; (2) provide education to pregnant women onantenatal care, PMTCT, birth, and postnatal care; (3) routinely follow up on all pregnant women to ascertain whether theyhave attended ANC; and (4) follow up on women who have missed ANC or PMTCT appointments.(Continued on next page)

* Correspondence: tbaernig@hsph.harvard.edu†Equal contributors1Department of Global Health and Population, Harvard School of PublicHealth, Huntington Avenue, Boston, Massachusetts 02115, USA8Wellcome Trust Africa Centre for Health and Population Studies, A2074Road, Mtubatuba, KwaZulu-Natal 3935, South AfricaFull list of author information is available at the end of the article

© 2014 Sando et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.

Sando et al. Trials 2014, 15:359 Page 2 of 13http://www.trialsjournal.com/content/15/1/359

(Continued from previous page)

Trial registration: ClinicalTrials.gov: EJF22802. Registration date: 14 May 2013.

Keywords: Study protocol, HIV, Antenatal care, Prevention of mother-to-child transmission, Community health workers,Uptake, Retention

BackgroundMother-to-child transmission of HIVThe Joint United Nations’ Programme on HIV/AIDS(UNAIDS) estimates that 260,000 children globally werenewly infected with HIV in the year 2012 [1], the vast ma-jority through vertical transmission of the virus frommother to child. Thus, even though the number of HIVinfections in children decreased by an estimated 52% be-tween 2001 and 2012, vertical transmission remains a sig-nificant problem, particularly in sub-Saharan Africa (SSA)[1]. Most of these infections are preventable. The WorldHealth Organization (WHO) estimates that in the absenceof any intervention the mother-to-child transmission(MTCT) rate is 15 to 45% [2]. Several studies in low-andmiddle-income countries have shown that by using mater-nal triple antiretroviral (ARV) regimens it is possible to re-duce the six-month MTCT rate to less than 5%, even inlow-resource settings (Table 1). The MTCT rate at sixmonths is often used as an outcome in studies on preven-tion of MTCT (PMTCT). However, if breastfeedingcontinues beyond six months of age, HIV can still betransmitted.

The obstacles of late and low antenatal care uptake andretentionWHO recommends the integration of PMTCT serviceswith maternal and child health services [10]. In inte-grated PMTCT services, antenatal care (ANC) is thekey entry point for PMTCT services (for example,through HIV testing). Therefore, in many low- andmiddle-income countries, one of the major challengesto achieving the maximum benefits from PMTCT ser-vices is late and low ANC attendance. The first ANCvisit is often delayed and women attend a low overallnumber of ANC visits [11,12]. WHO recommends thatpregnant women have their first ANC visit in the firsttrimester of pregnancy and attend a total of at leastfour visits [13]. In Tanzania, an estimated 98% ofwomen attended at least one ANC visit for their mostrecent live birth [14]. However, only 43% made fourANC visits or more, and only 15% of women attendedANC during the first trimester of pregnancy(Additional file 1). Late and inconsistent attendance ofANC means that any pre-existing health problems(such as sexually transmitted infections and anemia)as well as problems arising during pregnancy (such asurinary tract infections, malaria, gestational diabetes,

and pre-eclampsia) are more likely to be detected late,thereby significantly increasing the risks of adverseoutcomes from these conditions to the health of themother and the newborn child [15]. Moreover, de-pending on the delay in the first ANC visit, HIV-infected women may not be initiated on ARVs at all ormay receive ARVs late in pregnancy, decreasing the ef-fectiveness and efficacy of PMTCT [7].

Community health worker interventions to improve ANCand PMTCT outcomesThis cluster randomized health systems implementa-tion study, called the Familia Salama (Kiswahili for‘safe family’) trial, will test the effectiveness, cost-effectiveness, feasibility, and acceptability of a commu-nity health worker (CHW) intervention in improvingANC and PMTCT outcomes. While there is consider-able evidence to show that CHW interventions cancontribute to child health improvements [16,17], theevidence for the effect of CHW interventions onreproductive health services (including ANC andPMTCT) is scant [16,18]. A systematic review byWHO on CHW interventions has found that for inter-ventions with a maternal health component “almost allof the CHW-driven interventional studies showed (…)a[n] improvement in perinatal and postpartum serviceutilization indicators” (WHO 2010) [16], but it notedthat few studies were conducted at large scale and nonehad employed an experimental design. In addition, itfound a “remarkable dearth of information” on thecost-effectiveness of CHW programs [16]. ConcerningPMTCT, the WHO review did not identify a singlestudy of a CHW intervention to increase PMTCTuptake and/or retention [16]. Nonetheless, a more re-cent systematic review by UNAIDS has identifiedseveral examples of successful programs that useCHWs to improve PMTCT uptake and retention. How-ever, none of these programs were evaluated in arandomized trial [19].The Familia Salama trial will thus provide much

needed evidence on the effectiveness of communityhealth workers (CHWs) in increasing ANC and PMTCTuptake and retention. The results of this study will beparticularly important as it is a large-scale randomizedhealth systems implementation trial, which also assessesthe acceptability, feasibility and cost-effectiveness of theCHW intervention.

Table 1 Studies conducted in low-and middle-income countries, which achieved a six-month mother-to-child HIVtransmission rate of ≤5%

Country Year Participants Maternal regimen Infantregimen

Transmission rate at 6months (95% confidence

interval)

Botswana [3] 2006-2008

730 breastfeedingwomen1

Triple ARVs from 28-34 weeks until cessation ofbreastfeeding2

sd-NVP +AZT for 4weeks

1.1% (0.5%-2.2%)

Rwanda [4] 2005-2007

Mothers of 227breastfed and 305formula-fed infants

Formula-feeding: triple ARVs from 28 weeks untilbirth. Breastfeeding: triple ARVs until cessation of

breastfeeding2

sd-NVP +AZT for 7days

Breastfed: 1.8%3 (0.7%-4.8%)

Formula-fed: 1.0%3

(0.3%–3.0%)

Mozambique [5] 2005-2007

341 breastfeedingmother-infant pairs

Triple ARVs from 15 weeks until cessation ofbreastfeeding2

sd-NVP +AZT for 7days

2.1%

Mozambique,Malawi, andTanzania [6]

2004-2006

809 formula-feedingwomen

Triple ARVs from 25 weeks until cessation ofbreastfeeding2

sd-NVP +AZT for 7days

2.7%

Burkina Faso,Kenya, SouthAfrica [7]

2005-2008

Mothers of 805 infants4,5 Group 1: triple ARVs from 28-36 weeks until cessa-tion of breastfeeding2,6

sd-NVP +AZT for 7days

Group 1: 4.9% (3.1%-7.6%)

Group 2: AZT during pregnancy + sd-NVP and AZTat onset of labor7,8

Group 2: 8.4% (6.0%-11.6%)

Kenya [8] 2003-2009

Mothers of 487breastfed infants

Triple ARVs from 34-36 weeks until cessation ofbreastfeeding2

sd-NVP 5.0% (3.4%-7.4%)

Tanzania [9] 2004-2006

Mothers of 441breastfed infants

Triple ARVs from 34 weeks until cessation ofbreastfeeding2

AZT + 3TCfor 7 days

5.0% (3.2%-7.0%)

3TC, lamivudine; ARV, antiretroviral drugs; AZT, zidovudine; NVP, nevirapine; sd-NVP, single dose of nevirapine at birth.1Of these 730 women, 560 had a CD4-count ≥200 cells/mm3 and 170 a CD4-count <200 cells/mm3 or an AIDS-defining illness.2Women were advised to have concluded complete cessation of breastfeeding by 6 to 7 months postpartum, depending on the study.3These are transmission rates at 9 months (not 6 months).478% of infants born to women in each arm of the study were ever breastfed.5This is the number of participants randomized at the beginning of the study.6349 infants’ mothers were provided with this regimen.7339 infants’ mothers were provided with this regimen.8After a protocol amendment in 2006, women in the AZT arm also received AZT + 3TC 7 days postpartum.

Sando et al. Trials 2014, 15:359 Page 3 of 13http://www.trialsjournal.com/content/15/1/359

WHO’s PMTCT Option A and Option BHIV can be transmitted from mother to child duringpregnancy or childbirth through blood, or postnatallythrough breast milk. Until the early 2000s, PMTCT trialsand WHO recommendations focused on preventing thetransmission of HIV during pregnancy, birth, and theearly postnatal transmission period. When WHO pub-lished updated PMTCT guidelines in 2000, randomizedclinical trials had not tested PMTCT regimens withARVs beyond seven days postpartum for the mother andsix weeks postpartum for the infant [20]. Transmissionthrough breast milk after the first few weeks of life, how-ever, remained a serious challenge. Given the over-whelming benefits of breastfeeding, the lack of safealternatives to human milk in resource-limited countries,and the documented risk of HIV transmission throughbreastfeeding, there was an urgent need to make breast-feeding by HIV-infected women safer by preventingpostnatal transmission of the virus [21,22].In this context, research started to focus on how ARVs

can be given to the mother or the infant in the

postpartum period to reduce transmission rates whileallowing the infant unrestricted access to breast milk.These studies showed promising results for prophylacticARVs administered to infants [9,22-26]. The scientificevidence from these studies, along with new evidence onsafe feeding practices for HIV-exposed infants, early in-fant diagnosis, and revised maternal ARV eligibility, ledto the updating of WHO’s PMTCT guidelines in 2010[10].In its 2010 guidelines, WHO recommended the imme-

diate initiation of lifelong triple ARV treatment for allHIV-infected pregnant women with a CD4 cell count of≤350 cells/mm3, or who are in WHO clinical stage IIIor IV of HIV disease irrespective of CD4 cell count[10]. All infants born to pregnant women on lifelongtriple ARV treatment should receive nevirapine (NVP)or zidovudine (AZT) for four to six weeks regardless offeeding mode. This is in contrast to the 2006 guide-lines, which recommended lifelong treatment for HIV-infected pregnant women if the CD4 cell count was<200 cell/mm3, the woman was in clinical stage IV, or

Sando et al. Trials 2014, 15:359 Page 4 of 13http://www.trialsjournal.com/content/15/1/359

the woman was in clinical stage III with a CD4 cellcount of ≤350 cells/mm [27].For pregnant women who are not eligible for lifelong

ARV treatment, the WHO 2010 guidelines recommendeither of two drug regimens for PMTCT. The recom-mended maximum duration of breastfeeding, regardlessof the PMTCT option used, is 12 months unless anutritionally adequate and safe diet cannot be providedwithout breast milk [10].Option A: AZT prophylaxis from as early as 14 weeks

of gestation until delivery. If the woman has received AZTfor less than four weeks, then NVP (single dose) + AZT +lamivudine (3TC) should be administered every 12 hoursduring labor and delivery, and AZT + 3TC for seven dayspostpartum. If the woman has received AZT for morethan four weeks, another option is to continue AZT every12 hours during labor and then stop ARV treatment.Breastfeeding infants should be given NVP daily until oneweek after stopping breastfeeding and non-breastfeedinginfants should be given NVP or NVP (single dose) + AZTfor four to six weeks after delivery.Option B: one of the following triple ARV regimens

from as early as 14 weeks of gestation until one weekafter stopping breastfeeding: (1) AZT + 3TC + lopinavir/ritonavir (LPV/r), (2) AZT + 3TC + abacavir, (3) AZT +3TC + efavirenz (EFV), (4) tenofovir disoproxil fumarate(TDF) + 3TC (or emtricitabine) + EFV. All infants re-gardless of feeding mode receive NVP or AZT for fourto six weeks.Both the WHO 2010 and 2013 guidelines found that

the available evidence does not show whether one of thetwo PMTCT options described above is more efficaciousthan the other [10,28]. The WHO 2010 guidelines (themost current guidelines available at the beginning of thetrial period) recommended that policymakers choose theoption more suited to their cultural context, health sys-tem, and economic constraints [10]. However, it is stillunclear which option will be more effective, cost-effective, feasible, and acceptable when implemented inreal-life public-sector health systems in SSA. For ex-ample, regarding Option B, questions concerning poten-tial adherence fatigue during the breastfeeding periodand harm to mothers and infants from prolonged expos-ure to ARVs still remain largely unanswered [29]. It isnot known how such potential disadvantages compare tothe operational drawbacks of Option A, such as the needfor CD4 counts to determine the type of regimen to beinitiated, the needed changes from antepartum to intra-partum and postpartum regimens, and the requirementfor extended NVP administration in breastfeeding infants[29,30]. Evidence, particularly from large-scale, random-ized health system trials, on these questions is needed forpolicymakers to design locally appropriate and optimalPMTCT policies.

WHO’s PMTCT Option B+PMTCT has been a dynamic field in the past and recentchanges in WHO’s policy recommendations have beenrapid and not without controversy [29]. This trial wasdesigned when the latest policy guidance document onPMTCT that WHO had published were the 2010PMTCT guidelines. In 2013, WHO published a newguideline on the use of ARV drugs for treating and pre-venting HIV infection. In these updated guidelines Op-tion A is no longer recommended [28]. In addition,WHO added the so-called Option B+ as an alternative toOption B. Under Option B+, all HIV-infected pregnantwomen are started on ARVs irrespective of CD4 countand as soon as they have been diagnosed as HIV-infected.Even though WHO no longer recommends Option A,

it is still widely used across SSA since the implementa-tion of new PMTCT guidelines lag behind the WHOrecommendations in many countries, and this trial willprovide useful evidence for PMTCT programs in manylow- and middle-income countries. In its 2013 guide-lines, WHO states that ‘research is needed to determinehow to optimize acceptability, adherence and retentionon antiretroviral therapy (ART) in pregnant and breast-feeding women’ (p. 104) [28]. Improving acceptabilityand retention in PMTCT care are central aims of the Fa-milia Salama trial’s CHW intervention. In addition, thetrial will provide useful information for those sub-Saharan African countries considering a transition fromOption A to Option B or Option B+. In this context, itis important to note that in this trial and for practicalpurposes Option B is very similar to Option B+, becauseOption B and Option B+ do not differ during the courseof pregnancy and the breastfeeding period. It is onlyonce a woman has stopped breastfeeding that she wouldeither stop (under Option B) or continue her ART forlife (under Option B+). Since the entire length of thisstudy is only 17 months, most women participating inthis trial will have received the same treatment in theOption B arms of the trial that they would have receivedunder the Option B+ policy. An important difference,nevertheless, lies in the woman’s expectation regardingthe duration for which she should take ARVs, whichmay have an influence on adherence and retention. Byworking with, and collecting data from clinics and hospi-tals in wards, which were required to change theirPMTCT regimen from Option A (the standard of carein Tanzania) to Option B for this trial, we will gain im-portant insights on the implementation challenges oftransitioning from Option A to Option B or B+ .

PMTCT in TanzaniaIn 2010, the MTCT rate in Tanzania was 25% [31], with77% of HIV-infected pregnant women [1] and 74% of

Sando et al. Trials 2014, 15:359 Page 5 of 13http://www.trialsjournal.com/content/15/1/359

HIV-exposed infants [32] receiving ARVs for PMTCT(Additional file 1).The government of Tanzania began offering PMTCT

services in 2000 [33]. These services have now been inte-grated into Tanzania’s reproductive and child health ser-vices and should, therefore, be provided by the clinicalstaff in ANC, labor and delivery, and postnatal care facil-ities. However, not all ANC facilities offer ARVs, inwhich case women are referred to the ART treatmentprogram. In June 2012, Tanzania adopted WHO’s 2010PMTCT guidelines and implemented Option A as thestandard of care. Apart from the treatment and prophy-laxis of pregnant women and their newborns with ARVs,Tanzania’s PMTCT program includes voluntary testingof pregnant mothers with pre- and post-test counselling,infant feeding counselling, and family planning. Inaddition, the program includes the testing of HIV-exposed infants using polymerase chain reaction (PCR)at four to six weeks and six months after delivery, and arapid HIV-test after complete cessation of breastfeeding,usually between 9 and 18 months after delivery [34].Tanzania has decided to implement lifelong triple

ARVs for all HIV-infected pregnant women (Option B+)in 2013 with a graduated rollout. Because of this deci-sion the trial management team, in accordance with theData Safety and Monitoring Board and the TanzanianMinistry of Health and Social Welfare, decided to stopthe trial in May 2014 rather than in January 2015.

ObjectivesThe aim of this study is to determine the effectiveness,cost-effectiveness, acceptability, and feasibility of the fol-lowing two interventions when implemented in thehealthcare system in Dar es Salaam, Tanzania: (1) aCHW intervention to increase ANC and PMTCT uptakeand retention, (2) WHO Option A versus B for PMTCT,and (3) the interaction between the CHW interventionand the two PMTCT options.

Methods/DesignStudy durationThe trial is being carried out over a period of 17 monthsfrom January 2013 to May 2014.

Study settingThe Familia Salama trial is being carried out in theKinondoni and Ilala districts, two of the three adminis-trative districts of the Dar es Salaam region of Tanzania.In 2012, 69% (2,995,660 out of 4,364,541) of Dar esSalaam’s population lived in these two districts [35]. Dares Salaam is highly urbanized and is Tanzania’s mostdensely populated region with 3,133 people per squarekilometer [35].

Similar to other regions in Tanzania, virtually allwomen in Dar es Salaam attend at least one ANC visit.However, the percentage of women who deliver at ahealth facility (90%), have knowledge of MTCT (71% ofwomen knew that drugs can reduce the risk of MTCT)and received HIV-counselling during ANC (85%) isconsiderably higher in Dar es Salaam than in otherregions of Tanzania [14,36]. Dar es Salaam also has thesecond highest HIV-prevalence (9% among women andmen aged between 15 and 49 years) of any region inTanzania [14].

Study designThe Familia Salama trial is a cluster-randomized con-trolled trial that uses a two-by-two factorial design(Figure 1). Together, the districts in which the trialtakes place have 60 wards. These wards were randomlyallocated to either the CHW intervention (36 wards)or the normal standard of care in Dar es Salaam (24wards). The two arms were then randomized again toreceiving either WHO Option A or Option B forPMTCT (Figure 1). The number of clusters in thearms varies to reflect the difference in the populationsizes in each ward.The study is carried out by Management and Develop-

ment for Health (MDH, Tanzania). The Harvard Schoolof Public Health provides technical assistance for thetrial of the CHW intervention. MDH is a Tanzanianorganization based in Dar es Salaam, which works inpartnership with Tanzania’s Ministry of Health and So-cial Welfare. It provides technical and financial supportto 50 HIV treatment sites, 17 tuberculosis clinics, and180 PMTCT outlets across Dar es Salaam.

The community health worker interventionThe CHW intervention employs two cadres of healthworkers that already exist in the public-sector healthsystem in Tanzania. The first cadre is CHWs. Thiscadre is locally called home-based carers, or HBC, butwe will continue to refer to them as community healthworkers, because the activities that these healthworkers carry out are similar to those carried out byCHWs in most other settings. The second cadre isnurses who are responsible for community outreachfrom clinical facilities and whose responsibility includesfollowing up with HIV-infected pregnant women in thecommunity when they miss a PMTCT appointment(see below). This cadre is locally referred to ascommunity-based healthcare workers, or CBHC, butwe will refer to them as community outreach nurses inthis article, because this cadre is based in clinicalfacilities rather than in the community.In wards randomized to the CHW intervention, the

study received from the municipalities: (1) an additional

60 wards

Community intervention

36 wards

WHO Option B

22 wards(36,550 pregnancies)1

WHO Option A

14 wards(35,622 pregnancies)1

Standard of care

24 wards

WHO Option B

14 wards(37,098 pregnancies)1

WHO Option A

10 wards(44,424 pregnancies)1

Figure 1 The randomization scheme of the Familia Salama trial.

Sando et al. Trials 2014, 15:359 Page 6 of 13http://www.trialsjournal.com/content/15/1/359

31 community outreach nurses to supplement the exist-ing cadre of 21 community outreach nurses, with one tothree nurses being assigned to each ward, and (2) acadre of 215 CHWs, who work in the neighborhood inwhich their home is located. One to three CHWs wereassigned to each street in the intervention wardsdepending on the size of the street.

Responsibilities of the community health workersWhile the CHWs exist as a cadre in the public-sectorhealth system in Tanzania, they are currently not car-rying out any specific activities to support ANC andPMTCT uptake and retention in the districts wherethe trial takes place. In the CHW intervention arms of thetrial they are employed to fulfill the following functions:(1) identify pregnant women at the community levelthrough home visits and refer them to ANC; (2) routinelyrevisit all identified pregnant women to check whetherthey have taken up ANC and to support first and futureANC uptake through information, education, and coun-selling; (3) visit pregnant women at home who havemissed scheduled ANC or PMTCT appointments; (4) pro-vide education to pregnant women and other householdmembers on maternal health, general health issues (suchas nutrition, immunization, and sanitation) and the im-portance of delivering at a healthcare facility, testing forHIV during ANC visits, PMTCT, breastfeeding, postnatalcare, seeking early and appropriate healthcare in case ofan illness during pregnancy. As is the standard in Dar esSalaam’s healthcare system, the CHWs receive a monthlystipend equivalent to $US35.

Following up with women who have missed an ANC orPMTCT appointmentIn control wards, only women who have missed aPMTCT appointment are tracked, while in wards ran-domized to the CHW intervention CHWs carry outhome visits to pregnant women who have missed eithera PMTCT or an ANC appointment. In addition, CHWs

in the intervention wards routinely revisit all pregnantwomen to check whether they have attended ANC, untilthe women have attended ANC for the first time.In the community intervention control areas, the com-

munity outreach nurses contact women once they areover seven days late for their ANC or PMTCT appoint-ment to encourage them to attend. Contact is firstattempted by phone, and subsequently by a home visit.In the wards receiving the CHW intervention, the

CHWs support the community outreach nurses in carry-ing out home visits to pregnant women. During the firstANC visit, the ANC nurse obtains the woman’s addressand a written description of where and how her homecan be found (a so-called map cue), as well as thewoman’s telephone number. The community outreachnurses visit each ANC facility in their ward weekly tocollect the addresses and telephone numbers of allwomen who are over seven days late for their ANC orPMTCT appointment. The community outreach nursescall these women to remind them of their appointment.If a community outreach nurse cannot reach a womanon the phone, s/he passes the information about thewoman on to a CHW who will subsequently try to visitthe woman. In addition, the community outreach nursescollect forms for referral to ANC from the CHWs andmatch them with the referral forms obtained fromwomen at the ANC facilities. The community outreachnurses pass those referral forms for which there is nocorresponding clinic visit (signifying that the womanwas referred to ANC but did not attend) back to theCHW for follow-up at the woman’s home.

Community sensitization activitiesAs part of the community health intervention, commu-nity sensitization activities are organized in interventionareas. The research team, along with project and munici-pal community health coordinators, provide informationon maternal health and ANC to local leaders and en-courages them to spread educational messages to their

Sando et al. Trials 2014, 15:359 Page 7 of 13http://www.trialsjournal.com/content/15/1/359

communities. The community leaders are the WardDevelopment Committee at the ward level and the Mtaa(Kiswahili for ‘neighborhood’) Executive Committee atthe neighborhood level.

The trial of WHO Option A versus Option BWHO Option A is the current standard of care forPMTCT in Tanzania [34]. In the wards allocated to OptionA, HIV-infected women who are not eligible for lifelongARV therapy receive AZT from 14 weeks of gestation (orat the first PMTCT visit, if it takes place after the 14th weekof pregnancy) until delivery. This is followed by 12-hourlyAZT+ 3TC during labor and delivery and for seven dayspostpartum. If the woman has taken AZT for less than fourweeks, an additional single dose of NVP is administered atthe onset of labor. HIV-exposed breastfeeding infants aregiven NVP syrup once a day until one week after stoppingbreastfeeding while non-breastfeeding infants are givenNVP syrup for six weeks after delivery. Tanzania’s PMTCTguidelines recommend that HIV-infected women breast-feed exclusively for the first six months and stop breast-feeding 12 months after birth unless an adequate dietcannot be provided to the infant without breast milk [34].In the wards allocated to WHO Option B, HIV-

infected women not eligible for lifelong ARVs receiveAZT + 3TC + EFV from the 14th week of pregnancy untilone week after the complete cessation of breastfeeding.Regardless of the feeding option, all HIV-exposed infantsin the Option B wards receive NVP syrup for four to sixweeks after delivery.Both PMTCT options are provided by clinical staff in

ANC, labor and delivery, and postnatal care services.Some ANC facilities, however, do not provide ARVs, inwhich case the woman is referred to ART services.HIV PCR testing for infants is the standard of care in

Tanzania [34]. In all wards, PCR is performed at four tosix weeks and six months after delivery and a HIV rapiddiagnostic test (RDT) is administered after the completecessation of breastfeeding, usually between 9 and 18months of age. Infants’ CD4 count is measured every sixmonths as per the national guidelines. HIV-infectedpregnant women in all wards receive a daily double-doseof trimethoprim/sulfamethoxazole (Co-trimoxazole™),and exposed infants receive daily Co-trimoxazole™ syrupuntil their mother has stopped breastfeeding and theyare confirmed to be HIV-uninfected.

Patient and health worker satisfactionPatient satisfaction is a determinant of treatment uptake,adherence and retention, as well as an important healthsystems outcome in its own right [37,38]. Health workersatisfaction, on the other hand, is an important determin-ant of staff turnover [39], job-related motivation [40], andhealth worker migration [41]. It is plausible that the type

of PMTCT option and the CHW intervention have effectson patient and health worker satisfaction. Option B re-quires more frequent visits to PMTCT facilities, whichmay reduce patient satisfaction and increase healthworkers’ workload and job dissatisfaction. At the sametime, Option B has been widely hypothesized to be sim-pler to carry out than Option A, because it entails onlyone treatment for all HIV-infected pregnant and breast-feeding women [28]. This operational simplicity couldconceivably improve patient and provider satisfaction.Similarly, the CHW intervention implies several changesthat could affect patient and provider satisfaction. Thehome-based visits might lead to better or worse patientevaluation of the performance of the public-sector ANCand PMTCT systems. Leading and managing CHWsmight improve job satisfaction among the communityoutreach nurses (since they now have greater spans of re-sponsibility and have gained human resources to care forpregnant women in the community), or it might reducejob satisfaction (for example, if the community outreachnurses feel inadequately prepared for managing largenumbers of CHWs).In the Familia Salama study, we thus assess the causal

effects of Option B (versus A) and the CHW interven-tion on patient and health worker satisfaction. To collectthese data, we administer a questionnaire to patientsand health workers in all four trial arms in the secondyear of the implementation of the interventions.

TrainingWe have provided training in clinical PMTCT servicesto the facility-based health workers participating in thisstudy to ensure that Option A and B are carried out asassigned through the trial. In addition, we have trainedthe CHWs and community outreach nurses in the deliv-ery of the CHW intervention (Additional file 2). Thetrainings were based on the Tanzanian National PMTCTTraining Curriculum and the Tanzanian Community-Based Care Curriculum.

Study endpointsThe primary endpoints of this study are: (1) the percent-age of pregnant women making at least four antenatalclinic visits (as recommended by WHO); (2) the percent-age of pregnant women delivering at a healthcare facility;(3) the percentage of HIV-infected women receivingPMTCT; (4) the percentage of HIV-exposed infants whoreceived a confirmatory HIV test by six weeks after the ces-sation of breastfeeding; and (5) the percentage of infantsborn to HIV-infected mothers who have acquired HIV by6 weeks after the complete cessation of breastfeeding.The secondary endpoints of this study are: (1) the per-

centage of pregnant women who were tested for HIVduring pregnancy or labor and delivery; (2) the number

Sando et al. Trials 2014, 15:359 Page 8 of 13http://www.trialsjournal.com/content/15/1/359

of weeks of gestation at which pregnant women attendANC for the first time; (3) the percentage of HIV-infected pregnant women who completed PMTCT; and(4) the percentage of HIV-exposed infants who receivedPMTCT.In addition to the designated primary and secondary

endpoints, we will assess the causal impact of the inter-ventions on a range of other endpoints, which capture thefollowing outcomes: cost-effectiveness, social acceptability,patient satisfaction, and health worker satisfaction.

Eligibility criteriaInclusion criteriaAll pregnant women who either attend ANC at any ofthe clinics included in this study or who are identifiedby the CHWs during the routine household visits aseligible for participation in this study.

Exclusion criteriaAny pregnant woman who is not able to provide verbalinformed consent (for ANC attendees) or writteninformed consent (for PMTCT attendees) is ineligible toparticipate in the study.

Randomization procedureThe unit of randomization is a ward. A ward in the Ilalaand Kinondoni districts consists of two to ten streets and6,742 to 106,946 inhabitants (mean: 49,928) [35]. The 60wards in the Ilala and Kinondoni districts were dividedinto five strata by type of facility (military, high level, orstandard level) and district (Ilala and Kinondoni). A ‘highlevel’ facility was defined as having the capacity to manageobstetric complications, including performing caesariansections, while a ‘standard level’ facility does not have thiscapacity.Based on prior empirical evidence, ANC and PMTCT

outcomes differ significantly by facility type and district.Thus, the sample was stratified by these characteristicsto ensure adequate representation of the differentfacilities in the final sample and to gain statistical effi-ciency. Within each stratum, the wards were sorted indescending order of expected number of pregnancies,based upon the number of pregnancies reported inAugust 2011. A block randomization was then applied.The first four wards in each stratum were randomlyassigned to the four arms of the study. After each subse-quent round of four assignments, the assignment prob-abilities were revised to increase the probability that thearms with lower numbers of pregnancies would be morelikely to be chosen, thereby achieving the best balanceacross arms. This algorithm was repeated 500 times andthe run giving the greatest balance in the expected num-ber of pregnancies across the arms was selected.

While pregnant women can change the ANC facilitythey attend over the course of their pregnancy, clinicianswere advised to maintain women on the PMTCT optionon which they were started. Randomization into theCHW intervention is based on the ward in which thewoman lives.

Sample sizeWe expect that approximately 150,000 pregnancies wouldoccur in the two districts during the study period, includ-ing 10,500 pregnancies among HIV-infected women. Thisfigure is adjusted for the expected number of dropouts inthe current standard of care, as it was estimated using datafrom the participating facilities’ ANC registration booksfrom 2011. We calculated the minimum detectable differ-ence between the control and the intervention arms forthe primary endpoints, using methods for cluster-randomized trials [42] with unequal cluster sizes [40], withintra-cluster correlations between 0.001 and 0.030, as re-ported from similar cluster-randomized studies in low-and middle-income countries [43,44]. Primary outcomerates were taken from Tanzanian national data and studiesconducted in Botswana [45] and Zimbabwe [46]. Theminimum effect sizes that can be detected with a Type Ierror of 0.05 are modest and well within the range of whatcould be expected from this study.

Data collectionThe majority of the study data is collected from clinicalregisters that are in routine use in the Tanzanian health-care system (Additional file 3). However, we have intro-duced five additional registers to record additional datafor the purposes of this study. Data from all of these reg-isters is entered by specialized data entry personnel intoan electronic database written in Microsoft Access. Adata quality manager checks for data entry error anddata inconsistencies on an ongoing basis. In addition tothe assessment of endpoints for the final study analysis,the data is further used to monitor study progress duringweekly review meetings throughout the study period.The study team runs quality checks on all backups re-

ceived from the data entry personnel. These include thepercentage of missing values for the variables in the regis-ter and standard error and plausibility checks. In addition,the number of women entered in each clinical register iscompared to the monthly aggregate number reported bythe district level authorities. Any inconsistencies in thedata are discussed with individual data entry personneland, if necessary, followed up on with visits to the health-care facility and a review of the clinical register inquestion.In addition, MDH has trained 30 nurses to mentor

nurses at two to five other healthcare facilities. Clinicalmentors visit their mentees’ facilities at least once a week.

Sando et al. Trials 2014, 15:359 Page 9 of 13http://www.trialsjournal.com/content/15/1/359

Mentorship activities focus on the accurate and completefilling out of clinical registers and ensuring the availabilityof HIV test kits and ARV drugs. The clinical mentorshipprogram has been implemented in all wards in this trial.

Economic evaluationBased on our understanding of clinical activities,PMTCT Options A and B/B+ require different types ofresources and different resource levels, but the real-lifecost-effectiveness of implementing these various optionsin a setting like Dar-es-Salaam is not known. We willempirically measure resource utilization in all four inter-vention arms through a costing study. Health workertime, by cadre, for an average patient-case throughoutthe course of PMTCT will be estimated based on datafrom an empirical time-motion study. Study fieldworkers with medical training will follow PMTCTnurses, health officers, and doctors during randomly se-lected clinic days to measure task times per PMTCTpatient visit. These times will be used to estimatepatient-case times, using the data available in the studydatabase on the number and types of clinic visits per pa-tient over the course of PMTCT. Other resourceutilization, such as drugs and medical supplies, will beestimated based on standard clinical guidelines comple-mented by pharmaceutical and clinic records. Unit costswill be collected based on administrative data from thepublic-sector PMTCT program and procurement re-cords, including health worker salaries, drug prices,prices of diagnostic tests, and healthcare facility rents.

Data managementThe study team trained a cadre of 20 data entry personnelin data entry and management. These data entrypersonnel work at the larger healthcare facilities in thisstudy where desktop computers have been installed fordata entry. When a clinical register is not in use, the dataentry personnel enter the data from the register into aMicrosoft Access database on the desktop computer.There is a separate password-protected Microsoft Accessdatabase for each register. In addition, some data entrystaff form a mobile team that visits smaller healthcare fa-cilities and enters data from the registers into a databaseon a laptop. To merge the data with the central database,the same data personnel take a password-protectedbackup of the clinic database on a laptop to the MDHheadquarters at regular intervals (daily to monthly,depending on the size of the facility).

Analysis planThe data will be analyzed by intention-to-treat analysis atthe participant level. Thus, participants will be assigned tothe cluster in which they were resident at the start of thetrial regardless of whether they move to other clusters or

out of the trial areas during the study period. We will,however, still be able to determine the number of partici-pants who crossed over between healthcare facilities orare lost to follow-up by linking women across clinic visitsand clinical registers. To evaluate the effect of the inter-ventions on the primary endpoints of this program, wewill fit robust clustered log-binomial models [47]. Themain primary endpoint of this trial is the percentage of in-fants born to HIV-infected mothers who have acquiredHIV by 6 weeks after complete cessation of breastfeeding.Thus, the primary analysis will be run for this endpoint;other endpoints will be assessed in secondary analyses.The robust score test will be used to assess the statisticalsignificance of intervention effects and the 95% confidenceintervals will be based on the robust variance. In this way,the validity of inferences will not depend on the validity ofassumptions about the correlation structure within or be-tween wards, or on assumptions of normality or homosce-dasticity of generalized regression residuals.On average, randomization eliminates confounding;

hence, the primary analysis will be unadjusted for poten-tial confounding variables. However, because power canoften be improved by covariate adjustment, secondaryanalyses will adjust for potential confounding for othermeasured individual-level risk factors such as age, aswell as ward-level variables such as population size. Toassess the statistical significance of effect modification,the likelihood ratio test will be used, comparing modelswith main effects only to models that include relevantcross-product terms. When significant effect modifica-tion is detected, point and interval estimates of interven-tion effects stratified by the effect modifier will bepresented. Because effect modifiers are not hypothesizeda priori, the study is not powered to detect effect modifi-cation and we will cautiously interpret any post hoc ef-fect modification observed. All analysis will beconducted using the SAS statistical software package(SAS Institute, Cary, North Carolina).

Interim analyses and stopping rulesThe aggregate data (i.e., data not separated by assignmentto intervention or control arms) for each endpoint will beanalyzed continuously during this trial to monitor dataquality. In addition, an independent Data Safety and Mon-itoring Board (DSMB) has been formed and meets bian-nually to review data quality and completeness, progresson the trial, adherence to the protocol, and the safety ofparticipants. The board includes two Tanzanian and twoUnited States-based scientists. In closed session, the studyteam’s statistician provides the DSMB with data on eachendpoint disaggregated by assignment to intervention orcontrol arms. No other research team members have ac-cess to this report.

Sando et al. Trials 2014, 15:359 Page 10 of 13http://www.trialsjournal.com/content/15/1/359

There are no stopping rules for this trial because OptionB is the PMTCT option recommended by WHO [28], andbecause it is unlikely that the CHW intervention couldlead to adverse effects. However, safety endpoints arereviewed alongside the primary and secondary endpointsat each DSMB meeting. In addition, the DSMB was askedto review the trial protocol as new PMTCT guidelines,such as the WHO guidelines, are published. The DSMB,therefore, reviewed the study protocol when the newWHO 2013 guidelines were published and decided againstamending the study design or discontinuing the trial.

Ethical issuesApprovalsThe National Institutes of Medical Research in Tanzaniagranted ethical approval for the study in June 2012. Theinstitutional review board of the Harvard School of Pub-lic Health approved the analysis of the trial of the CHWintervention in January 2013; this study received con-tinuing ethics approval in February 2014.

ConsentVerbal consent is obtained in the CHW intervention attwo time points: (1) by the CHWs before they enter ahousehold for a counselling and information session;and (2) by the healthcare worker (a nurse or physician)at the first ANC visit. In the trial of WHO Option Aversus Option B, the PMTCT health workers elicitwritten informed consent from HIV-infected pregnantwomen during their first PMTCT visit.

ConfidentialityParticipants’ confidentiality is protected several processes.Firstly, for the information sessions provided during

home visits, the CHW and the pregnant woman willjointly identify a room in the house or place outside thehouse that allows for counselling in private. Secondly,to reduce the risk of accidental disclosure of a woman’sHIV status, the procedures in the CHW interventionare the same for HIV-uninfected as for HIV-infectedpregnant women. Thirdly, database access is restrictedto a limited number of study staff for data entry andextraction. In addition, all staff involved in the studyhave been trained before initiation of the study toensure that they are aware of the rules regarding confi-dentiality and data protection. All computers that areused for data entry as part of this study are protectedby an individual password known only to the data entryclerk. In addition, the study database on each computer,where data from particular ANC and PMTCT clinics isstored, is protected by an individual password. AtMDH, all computers and the study database areprotected by passwords. Only individuals who arephysically present at the MDH facility in Dar es Salaam

can access the central server database, i.e., it is notpossible to gain access from outside the facility. At anypoint in time, only 5-7 specifically authorized dataentry staff at MDH have access to data in the centralserver database. No other study researchers have accessto de-identified data in the central database.

Illness of participantsWhen CHWs detect illness in a family member duringtheir home-based visits, they will encourage referral toan appropriate health facility to the best of their ability.

Sustainability and scalabilityThis trial will provide critical information on the scal-ability and sustainability of WHO Option A versus Band on a CHW intervention to improve ANC andPMTCT uptake and retention. It achieves this by inte-grating the interventions directly into the public-sectorhealthcare system rather than building a parallel struc-ture for the delivery of the trial interventions, and byassessing feasibility, acceptability, and cost-effectiveness,all of which are crucial to sustainability and scalability.

Dissemination of findingsThe Tanzanian Ministry of Health and Social Welfare(MHSW) has been involved in the planning of this studyfrom the start. In addition, three MHSW staff are mem-bers of the study team. This collaboration will ensurethe dissemination of our findings to the MHSW on acontinuous basis through both informal (such as per-sonal meetings) and formal channels (such as studymeetings and reports). In addition, the results of thestudy will be disseminated to the respective participatingwards, antenatal clinics, and health workers through per-sonal meetings and reports. Globally, the results of thestudy will be published in academic journals and pre-sented at relevant conferences.

Health systems implementation trialThis study can be categorized as a health systems imple-mentation trial in contrast to other types of trials suchas clinical trials. The study is a health systems trial forseveral reasons:For one, the study is implemented through the public-

sector health system in two of the three districts of Dares Salaam. The trial has only become possible becausethe Tanzanian MHSW and the National Institute ofMedical Research (NIMR) have supported it. The inter-ventions themselves are implemented through thepublic-sector health systems structures and processes inDar es Salaam, using only health worker cadres thatalready exist in the Tanzanian public-sector healthsystem.

Sando et al. Trials 2014, 15:359 Page 11 of 13http://www.trialsjournal.com/content/15/1/359

Option B has been introduced through educating exist-ing health workers in all the facilities in the randomly se-lected wards in the two study districts about this PMTCToption. The trial team continues to engage on an ongoingbasis with the health workers providing PMTCT and ANCin all four arms of the trial to ensure high fidelity in imple-menting the different intervention exposures.In addition, the outcomes are assessed using the real-

life public-sector health system’s data collection systems.The trial team supports the routine data collection in-cluding the extraction of a wide range of routinely col-lected data into an electronic database and data qualitycontrol. While a few additional variables are collectedspecifically for this trial, most of the primary andsecondary outcomes of interest are collected throughthe routine data collection mechanisms in place in theTanzanian public-sector health system, irrespective ofthis study (see Additional file 3).Finally, all pregnant women who seek ANC and

PMTCT services in the public-sector health system inthe Kinondoni and Ilala districts of Dar es Salaam areoffered to participate in the study. Unlike in clinicaltrials, only a few selection criteria apply and the follow-up of participants regarding exposure and outcomes iscarried out largely through the public-sector healthsystem and linkage of individuals across differentroutinely collected data forms.The study is an implementation trial because it assesses

(1) the effectiveness of PMTCT Option A and B and theCHW intervention under the conditions of a real-lifehealth system in SSA that has performed imperfectly inthe past in ensuring universal PMTCT coverage of HIV-infected pregnant women; (2) how the effects of PMTCTOption B are modified by a CHW intervention; (3) out-comes other than effectiveness that are essential to informhealth policy and practice, including cost-effectiveness,technical feasibility, and the social acceptability of OptionB and the CHW intervention; and (4) the effects of OptionB and the CHW intervention on implementationoutcomes such as patient satisfaction, quality of care, andhealth worker job satisfaction. In addition, the trialattempts to affect PMTCT-related outcomes through aCHW intervention that focuses on ensuring early andcomprehensive uptake of ANC (the initial health systemfunction that needs to perform well in order to detectHIV-infected pregnant women and to link them success-fully to PMTCT), and good retention in ANC andPMTCT through home visits of pregnant women tosupport continued ANC and PMTCT attendancethroughout the course of pregnancy.

Trial statusThe study is funded by Comic Relief - UK and the EltonJohn AIDS Foundation. The National Institute of

Medical Research in Tanzania approved the study in July2012 and the institutional review board of the HarvardSchool of Public Health approved the trial of the CHWintervention in January 2013. Randomization, recruitmentof project staff, CHWs and additional community out-reach nurses, and training of healthcare workers has beencompleted. Enrollment in the trial commenced in January2013. The trial will be stopped in May 2014. The trial isregistered at ClinicalTrials.gov under the registrationnumber EJF22802.

Additional files

Additional file 1: HIV/AIDS, ANC, birth attendance, and MTCT inTanzania.

Additional file 2: The facility- and community-based trainings inthe Familia Salama trial.

Additional file 3: List of data sources for the Familia Salama trial.

Abbreviations3TC: Lamivudine; ANC: Antenatal care; ARV: Antiretroviral drug;ART: Antiretroviral therapy; AZT: Zidovudine; CBHC: Community-basedhealthcare worker; CD4: Cluster of Differentiation 4; EFV: Efavirenz;HCW: Healthcare Worker; HIV: Human Immunodeficiency Virus; LPV/r: Lopinavir/Ritonavir; MDH: Management and Development for Health;MHSW: Ministry of Health and Social Welfare; MTCT: Mother-to-child HIVtransmission; MTUHA: Tanzanian health management information system;NIMR: National Institute of Medical Research; NVP: Nevirapine;PCR: Polymerase Chain Reaction; PMTCT: Prevention of mother-to-child HIVtransmission; RDT: Rapid Diagnostic Test; SSA: Sub-Saharan Africa;TDF: Tenofovir Disoproxil Fumarate; UNAIDS: Joint United NationsProgramme on HIV/AIDS; WHO: World Health Organization.

Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsDS (Sando) and PG contributed equally to this manuscript. DS (Sando), TWB,PG, GC, WWF and DS (Spiegelman) conceptualized the study and thismanuscript. PG, TWB and DS wrote the first draft of the paper. LM (Magesa),IA, LM (Machumi), MM, NL, DS (Spiegelman), EM, HS, PM, HN, GC, CK, AME,WWF and TWB commented on, and edited drafts of the paper. GC and TWBare principal investigators of the study. DS (Sando), LM, IA, LM, EM and MMcoordinate, manage, and implement the trial. All authors read and approvedthe final manuscript.

AcknowledgementsThe study is funded by Comic Relief - UK and the Elton John AIDSFoundation.

Author details1Department of Global Health and Population, Harvard School of PublicHealth, Huntington Avenue, Boston, Massachusetts 02115, USA.2Management and Development for Health, Mwai Kibaki Road, Dar esSalaam, Tanzania. 3Departments of Epidemiology and Biostatistics, HarvardSchool of Public Health, Huntington Avenue, Boston, Massachusetts 02115,USA. 4School of Public Health, Muhimbili University of Health and AlliedSciences, United Nations Road, Dar es Salaam, Tanzania. 5School of Medicine,Muhimbili University of Health and Allied Sciences, United Nations Road, Dares Salaam, Tanzania. 6Department of Public Health Sciences,Tomtebodavägen, Karolinska Institutet, Solna, SE-171 77 Stockholm, Sweden.7Department of Infectious Diseases, Karolinska University Hospital,Karolinskavägen, Solna, SE-171 76 Stockholm, Sweden. 8Wellcome TrustAfrica Centre for Health and Population Studies, A2074 Road, Mtubatuba,KwaZulu-Natal 3935, South Africa.

Sando et al. Trials 2014, 15:359 Page 12 of 13http://www.trialsjournal.com/content/15/1/359

Received: 17 October 2013 Accepted: 23 May 2014Published: 15 September 2014

References1. UNAIDS: Global Report - UNAIDS Report on the Global AIDS Epidemic 2013. Geneva:

UNAIDS; 2013 [http://www.unaids.org/en/media/unaids/contentassets/documents/epidemiology/2013/gr2013/UNAIDS_Global_Report_2013_en.pdf]

2. World Health Organization: Mother-to-Child Transmission of HIV. Geneva:World Health Organization; 2014. [http://www.who.int/hiv/topics/mtct/en/]

3. Shapiro RL, Hughes MD, Ogwu A, Kitch D, Lockman S, Moffat C, MakhemaJ, Moyo S, Thior I, McIntosh K, van Widenfelt E, Leidner J, Powis K,Asmelash A, Tumbare E, Zwerski S, Sharma U, Handelsman E, Mburu K,Jayeoba O, Moko E, Souda S, Lubega E, Akhtar M, Wester C, Tuomola R,Snowden W, Martinez-Tristani M, Mazhani L, Essex M: Antiretroviralregimens in pregnancy and breast-feeding in Botswana. N Engl J Med2010, 362:2282–2294.

4. Peltier CA, Ndayisaba GF, Lepage P, van Griensven J, Leroy V, Pharm CO,Ndimubanzi PC, Courteille O, Arendt V: Breastfeeding with maternalantiretroviral therapy or formula feeding to prevent HIV postnatalmother-to-child transmission in Rwanda. AIDS 2009, 23:2415–2423.

5. Marazzi MC, Nielsen-Saines K, Buonomo E, Scarcella P, Germano P, Majid NA,Zimba I, Ceffa S, Palombi L: Increased infant human immunodeficiencyvirus-type one free survival at one year of age in sub-saharan Africa withmaternal use of highly active antiretroviral therapy during breast-feeding.Pediatr Infect Dis J 2009, 28:483–487.

6. Palombi L, Marazzi MC, Voetberg A, Magid NA: Treatment accelerationprogram and the experience of the DREAM program in prevention ofmother-to-child transmission of HIV. AIDS 2007, 21(Suppl 4):S65–S71.

7. De Vincenzi I: Triple antiretroviral compared with zidovudine andsingle-dose nevirapine prophylaxis during pregnancy and breastfeedingfor prevention of mother-to-child transmission of HIV-1 (Kesho Borastudy): a randomised controlled trial. Lancet Infect Dis 2011, 11:171–180.

8. Thomas TK, Masaba R, Borkowf CB, Ndivo R, Zeh C, Misore A, Otieno J,Jamieson D, Thigpen MC, Bulterys M, Slutsker L, De Cock KM, Amornkul PN,Greenberg AE, Fowler MG: Triple-antiretroviral prophylaxis to preventmother-to-child HIV transmission through breastfeeding–the KisumuBreastfeeding Study, Kenya: a clinical trial. PLoS Med 2011,8:e1001015.

9. Kilewo C, Karlsson K, Ngarina M, Massawe A, Lyamuya E, Swai A, LipyogaR, Mhalu F, Biberfeld G: Prevention of mother-to-child transmission ofHIV-1 through breastfeeding by treating mothers with tripleantiretroviral therapy in Dar es Salaam, Tanzania: the Mitra Plusstudy. J Acquir Immune Defic Syndr 2009, 52:406–416.

10. World Health Organization: Antiretroviral Drugs for Treating Pregnant Womenand Preventing HIV Infection in Infants. Recommendations for a Public HealthApproach. 2010 Version. Geneva: World Health Organization; 2010 [http://whqlibdoc.who.int/publications/2010/9789241599818_eng.pdf]

11. World Health Organization: World Health Statistics. A Snapshot of GlobalHealth. 2012. Geneva: World Health Organization; 2012 [http://www.who.int/gho/publications/world_health_statistics/2012/en/]

12. World Health Organization: World Health Statistics 2013. Geneva: WorldHealth Organization; 2013 [http://apps.who.int/iris/bitstream/10665/81965/1/9789241564588_eng.pdf?ua=1]

13. World Health Organization: Antenatal Care Randomized Trial: Manual for theImplementation of the New Model. Geneva: World Health Organization; 2002.[http://whqlibdoc.who.int/hq/2001/WHO_RHR_01.30.pdf]

14. Tanzania Commission for AIDS Zanzibar AIDS Commission, National Bureau ofStatistics, Office of Chief Government Statistician, Macro International Inc:Tanzania HIV/AIDS and Malaria Indicator Survey 2007–08. Dar es Salaam:National Bureau of Statistics; 2008 [http://www.nbs.go.tz/tnada/index.php/catalog/9]

15. World Health Organization: Antenatal Care in Developing Countries – Promises,Achievements and Missed Opportunities. Geneva: World Health Organization;2003 [http://www.childinfo.org/files/antenatal_care.pdf]

16. World Health Organization: Global Experience of Community Health Workersfor Delivery of Health Related Millennium Development Goals: A SystematicReview, Country Case Studies, and Recommendations for Integration into NationalHealth Systems. Geneva: World Health Organization; 2010. [http://www.who.int/workforcealliance/knowledge/resources/chwreport/en/]

17. Lewycka S, Mwansambo C, Rosato M, Kazembe P, Phiri T, Mganga A, Chapota H,Malamba F, Kainja E, Newell M-L, Greco G, Pulkki-Brännström A-M,

Skordis-Worrall J, Vergnano S, Osrin D, Costello A: Effect of women’s groupsand volunteer peer counselling on rates of mortality, morbidity, and healthbehaviours in mothers and children in rural Malawi (MaiMwana): a factorial,cluster-randomised controlled trial. Lancet 2013, 381:1721–1735.

18. Viswanathan K, Hansen PM, Rahman MH, Steinhardt L, Edward A, Arwal SH,Peters DH, Burnham G: Can community health workers increase coverageof reproductive health services? J Epidemiol Community Health 2012,66:894–900.

19. UNAIDS: Promising Practices in Community Engagement for Elimination ofNew HIV Infections among Children by 2015 and Keeping Their Mothers Alive.Geneva: UNAIDS; 2012 [http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2012/20120628_JC2281_PromisingPracticesCommunityEngagements_en.pdf]

20. World Health Organization: New Data on the Prevention of Mother-to-ChildTransmission of HIV and Their Policy Implications: Conclusions andRecommendations. Geneva: World Health Organization; 2000 [http://whqlibdoc.who.int/hq/2001/WHO_RHR_01.28.pdf].

21. Fowler MG, Newell ML: Breast-feeding and HIV-1 transmission inresource-limited settings. J Acquir Immune Defic Syndr 2002, 30:230–239.

22. Shetty AK, Maldonado Y: Preventing mother-to-child transmission of humanimmunodeficiency virus type 1 in resource-poor countries. Pediatr Infect DisJ 2003, 22:553–555.

23. Bedri A, Gudetta B, Isehak A, Kumbi S, Lulseged S, Mengistu Y, Bhore AV,Bhosale R, Varadhrajan V, Gupte N, Sastry J, Suryavanshi N, Tripathy S, MmiroF, Mubiru M, Onyango C, Taylor A, Musoke P, Nakabiito C, Abashawl A,Adamu R, Antelman G, Bollinger RC, Bright P, Chaudhary MA, Coberly J,Guay L, Fowler MG, Gupta A, Hassen E et al: Extended-dose nevirapine to6 weeks of age for infants to prevent HIV transmission via breastfeedingin Ethiopia, India, and Uganda: an analysis of three randomisedcontrolled trials. Lancet 2008, 372:300–313.

24. Kilewo C, Karlsson K, Massawe A, Lyamuya E, Swai A, Mhalu F, Biberfeld G:Prevention of mother-to-child transmission of HIV-1 throughbreast-feeding by treating infants prophylactically with lamivudine inDar es Salaam, Tanzania: the MITRA Study. J Acquir Immune Defic Syndr2008, 48:315–323.

25. Kumwenda NI, Hoover DR, Mofenson LM, Thigpen MC, Kafulafula G, Li Q,Mipando L, Nkanaunena K, Mebrahtu T, Bulterys M, Fowler MG, Taha TE:Extended antiretroviral prophylaxis to reduce breast-milk HIV-1transmission. N Engl J Med 2008, 359:119–129.

26. Chasela CS, Hudgens MG, Jamieson DJ, Kayira D, Hosseinipour MC, Kourtis AP,Martinson F, Tegha G, Knight RJ, Ahmed YI, Kamwendo DD, Hoffman IF, EllingtonSR, Kacheche Z, Soko A, Wiener JB, Fiscus SA, Kazembe P, Mofolo IA,Chigwenembe M, Sichali DS, van der Horst CM: Maternal or infant antiretroviraldrugs to reduce HIV-1 transmission. N Engl J Med 2010, 362:2271–2281.

27. World Health Organization: Antiretroviral Drugs for Treating Pregnant Womenand Preventing HIV Infection in Infants in Resource-Limited Settings, TowardsUniversal Access. Geneva: World Health Organization; 2006. [http://www.who.int/hiv/pub/guidelines/WHOPMTCT.pdf]

28. World Health Organization: Consolidated Guidelines on the Use ofAntiretroviral Drugs for Treating and Preventing HIV Infection. 2013 Version.Geneva: World Health Organization; 2013 [http://apps.who.int/iris/bitstream/10665/85321/1/9789241505727_eng.pdf]

29. Van de Perre P, Tylleskär T, Delfraissy J-F, Nagot N: How evidence basedare public health policies for prevention of mother to child transmissionof HIV? BMJ 2013, 346:f3763.

30. Ngarina M, Tarimo EAM, Naburi H, Kilewo C, Mwanyika-Sando M, Chalamilla G,Biberfeld G, Ekstrom AM: Women's Preferences Regarding Infant or MaternalAntiretroviral Prophylaxis for Prevention of Mother-To-Child Transmissionof HIV during Breastfeeding and Their Views on Option B+ in Dar esSalaam, Tanzania. PLoS ONE 2014, 9(1):e85310.

31. World Health Organization, UNAIDS, UNICEF: Global HIV/AIDS Response,Epidemic Update and Health Sector Progress towards Universal Access.Geneva: World Health Organization; 2011. [http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2011/20111130_UA_Report_en.pdf]

32. UNICEF: Towards an AIDS-Free Generation. Children and AIDS: SixthStocktaking Report, 2013. New York City: UNICEF; 2013. [http://www.childrenandaids.org/files/str6_full_report_29-11-2013.pdf]

33. Ministry of Health and Social Welfare: National Scale up Plan for PMTCT andPediatric HIV Care, Treatment and Support. Dar es Salaam: Ministry of Healthand Social Welfare; 2008.

Sando et al. Trials 2014, 15:359 Page 13 of 13http://www.trialsjournal.com/content/15/1/359

34. Ministry of Health and Social Welfare: Prevention of Mother to ChildTransmission of HIV: National PMTCT Guidelines. Dar es Salaam: Ministry ofHealth and Social Welfare; 2011.

35. National Bureau of Statistics: 2012 Population and Housing Census. Dar esSalaam: National Bureau of Statistics; 2012 [http://www.nbs.go.tz/]

36. National Bureau of Statistics: Tanzania Demographic and Health Survey 2010.Dar es Salaam: National Bureau of Statistics; 2010 [http://www.nbs.go.tz/takwimu/references/2010TDHS.pdf]

37. Roberts MJ, Hsiao W, Berman P, Reich MR: Getting Health Reform Right: AGuide to Improving Performance and Equity. New York: Oxford UniversityPress; 2004.

38. Kravitz R: Patient satisfaction with health care: critical outcome or trivialpursuit? J Gen Intern Med 1998, 13:280–282.

39. Lu H, While AE, Barriball KL: Job satisfaction among nurses: a literaturereview. Int J Nurs Stud 2005, 42:211–227.

40. Eldridge SM, Ashby D, Kerry S: Sample size for cluster randomized trials:effect of coefficient of variation of cluster size and analysis method.Int J Epidemiol 2006, 35:1292–1300.

41. World Health Organization: Improving Health Worker Performance: In Searchof Promising Practices. Geneva: World Health Organization; 2006[http://www.who.int/hrh/resources/improving_hw_performance.pdf]

42. Donner A, Klar N: Design and Analysis of Cluster Randomization Trials inHealth Research. London, New York: Arnold and Oxford University Press;2000.

43. Feldblum PJ, Chen-Mok M, Bwayo JJ, Omari M, Kuyoh M, Ryan KA:Intracluster correlation of STD prevalence in a community interventiontrial in Kenya. Lancet 1999, 354:1356–1357.

44. Jokhio AH, Winter HR, Cheng KK: An intervention involving traditionalbirth attendants and perinatal and maternal mortality in Pakistan.N Engl J Med 2005, 352:2091–2099.

45. Creek TL, Ntumy R, Seipone K, Smith M, Mogodi M, Smit M, Legwaila K,Molokwane I, Tebele G, Mazhani L, Shaffer N, Kilmarx PH: Successfulintroduction of routine opt-out HIV testing in antenatal care inBotswana. J Acquir Immune Defic Syndr 2007, 45:102–107.

46. Ciaranello AL, Perez F, Maruva M, Chu J, Engelsmann B, Keatinge J,Walensky RP, Mushavi A, Mugwagwa R, Dabis F, Freedberg KA: WHO 2010guidelines for prevention of mother-to-child HIV transmission inZimbabwe: modeling clinical outcomes in infants and mothers.PLoS One 2011, 6:e20224.

47. Fitzmaurice JM, Laird NM, Ware JH: Applied Longitudinal Analysis. Hoboken,New Jersey: John Wiley and Sons; 2004.

doi:10.1186/1745-6215-15-359Cite this article as: Sando et al.: Evaluation of a community healthworker intervention and the World Health Organization’s Option Bversus Option A to improve antenatal care and PMTCT outcomes in Dares Salaam, Tanzania: study protocol for a cluster-randomized controlledhealth systems implementation trial. Trials 2014 15:359.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit